五角大樓與OpenAI、谷歌和輝達簽署機密AI協議
五角大樓與OpenAI、谷歌、微軟、亞馬遜、輝達、xAI及Reflection簽署了機密AI協議。這些協議允許合法作戰使用，旨在建立一支AI優先的戰鬥部隊。Anthropic因拒絕在大規模監控和全自動武器方面放鬆底線而被排除在外，並被列為供應鏈風險，儘管此前曾有2億美元的交易。
The Defense Department secured contracts allowing seven companies—OpenAI, Google, Microsoft, Amazon, Nvidia, xAI, and Reflection—to provide AI tools for classified military use. Defense Department CTO Emil Michael stated these agreements aim to establish an 'AI-first fighting force.' The contracts formalize recent 'lawful use' understandings with OpenAI and xAI, while Google reportedly finalized a similar deal through a separate report.
Anthropic was excluded from these agreements and labeled a supply-chain risk. This follows a dispute where Anthropic rejected terms regarding mass domestic surveillance and fully autonomous weapons. Despite holding a previous $200 million contract for classified work, Anthropic sued the federal government and secured a temporary injunction.